Search USGS⌕ Search

SEARCH · Search USGS

Search Search USGS

Search indexed USGS publications on groundwater, aquifers, geologic maps, mineral resources and earthquakes. Explore source records by subject and place.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

2 recordsLinked to original sources

Leveraging wildfire to augment forest management and amplify forest resilience

Successive catastrophic wildfire seasons in western North America have escalated the urgency around reducing fire risk to communities and ecosystems. In historically frequent-fire forests, fuel buildup as a result of fire exclusion is contributing to increased fire severity. The probability of high-severity fire can be reduced by active forest management that reduces fuels, prompting federal and state agencies to commit significant resources to increase the pace and scale of fuel reduction treatments. However, lower severity areas of wildfires also have the potential to act as “treatments,” and even catastrophic fires with large areas of high severity can still have substantial areas of lower severity fire that may be improving forest conditions locally. We quantified active management and wildfire severity across yellow pine and mixed conifer (YPMC) forests in the Sierra Nevada of California over a 22-year period (2001–2022). We did not detect increases in the area treated through time, but the area of beneficial wildfire (low to moderate severity) increased substantially, exceeding active treatment area in 8 of 22 years. Overall, beneficial wildfire treated ~17% more area than all treatments combined, and roughly four times more area than fire-related treatments alone. We then used disturbance history to evaluate resistance to high-severity wildfire and forest loss across the YPMC range. Of the 2.3 million ha YPMC of forests in 2001, 20% lost mature forests due to high-severity fire by 2022, which is nearly half of all YPMC area burned. Most of the landscape (47%) remains at risk of high-severity fire because it had no restorative disturbances, but 33% of the study area has some level of resistance to high-severity wildfire. In these areas, resistance will need to be enhanced and maintained over time via active management or managed wildfire, but these treatment needs will likely outpace capacity even under optimistic implementation scenarios. Given limited resources for implementing active management and the likelihood of a more fiery future, incorporating beneficial wildfire into landscape-level treatment planning has the potential to amplify the impact of active management treatments.

California, Nevada↗

Mapping bedrock outcrops in the Sierra Nevada Mountains (California, USA) using machine learning

Accurate, high-resolution maps of bedrock outcrops can be valuable for applications such as models of land–atmosphere interactions, mineral assessments, ecosystem mapping, and hazard mapping. The increasing availability of high-resolution imagery can be coupled with machine learning techniques to improve regional bedrock outcrop maps. In the United States, the existing 30 m U.S. Geological Survey (USGS) National Land Cover Database (NLCD) tends to misestimate extents of barren land, which includes bedrock outcrops. This impacts many calculations beyond bedrock mapping, including soil carbon storage, hydrologic modeling, and erosion susceptibility. Here, we tested if a machine learning (ML) model could more accurately map exposed bedrock than NLCD across the entire Sierra Nevada Mountains (California, USA). The ML model was trained to identify pixels that are likely bedrock from 0.6 m imagery from the National Agriculture Imagery Program (NAIP). First, we labeled exposed bedrock at twenty sites covering more than 83 km 2 (0.13%) of the Sierra Nevada region. These labels were then used to train and test the model, which gave 83% precision and 78% recall, with a 90% overall accuracy of correctly predicting bedrock. We used the trained model to map bedrock outcrops across the entire Sierra Nevada region and compared the ML map with the NLCD map. At the twenty labeled sites, we found the NLCD barren land class, even though it includes more than just bedrock outcrops, accounted for only 41% and 40% of mapped bedrock from our labels and ML predictions, respectively. This substantial difference illustrates that ML bedrock models can have a role in improving land-cover maps, like NLCD, for a range of science applications.

California↗